Re: The very latest emacs (25.1.50) segfault

[Eli Zaretskii <eliz@gnu.org>]

> Date: Mon, 11 Jan 2016 13:57:11 +0900
> From: Masaru Nomiya <nomiya@galaxy.dti.ne.jp>
> 
> Program received signal SIGSEGV, Segmentation fault.
> ftfont_shape_by_flt (matrix=0x3bd67d8, otf=<optimized out>, ft_face=0x3cbb300, 
>     font=0x3bd66e0, lgstring=12801653) at ftfont.c:2598
> 2598          int c1 = LGLYPH_CHAR (LGSTRING_GLYPH (lgstring, 1));
> (gdb) backtrace
> #0  ftfont_shape_by_flt (matrix=0x3bd4868, otf=<optimized out>, ft_face=
>     0x3d37310, font=0x3bd4770, lgstring=12801653) at ftfont.c:2598
> #1  ftfont_shape (lgstring=12801653) at ftfont.c:2690
> #2  0x00000000005b0aea in xftfont_shape (lgstring=12801653) at xftfont.c:664
> #3  0x00000000005608c5 in Ffont_shape_gstring (gstring=12801653) at font.c:4410
> #4  0x0000000000553570 in Ffuncall (nargs=2, args=<optimized out>)
>     at eval.c:2693
> #5  0x0000000000586235 in exec_byte_code (bytestr=31525964, vector=0, maxdepth=
>     62736488, args_template=140737488306080, nargs=2, args=0x2, args@entry=0x0)
>     at bytecode.c:880
> #6  0x0000000000552fc2 in funcall_lambda (fun=9765533, nargs=nargs@entry=5, 
>     arg_vector=arg_vector@entry=0x7fffffff41b8) at eval.c:2921
> #7  0x0000000000553383 in Ffuncall (nargs=nargs@entry=6, args=args@entry=
>     0x7fffffff41b0) at eval.c:2754
> #8  0x0000000000552001 in internal_condition_case_n (bfun=0x5531a0 <Ffuncall>, 
>     nargs=nargs@entry=6, args=args@entry=0x7fffffff41b0, 
>     handlers=handlers@entry=44448, hfun=hfun@entry=
>     0x43b180 <safe_eval_handler>) at eval.c:1389
> #9  0x000000000042b766 in safe__call (inhibit_quit=inhibit_quit@entry=false, 
>     nargs=nargs@entry=6, func=<optimized out>, ap=ap@entry=0x7fffffff4258)
>     at xdisp.c:2555
> #10 0x000000000043832f in safe_call (nargs=nargs@entry=6, func=<optimized out>)
>     at xdisp.c:2571
> #11 0x00000000005a2b0c in autocmp_chars (rule=rule@entry=19429861, 
>     charpos=charpos@entry=305, bytepos=322, limit=<optimized out>, limit@entry=
>     306, win=win@entry=0x1292a90, face=face@entry=0x3dca4e0, 
>     string=string@entry=0) at composite.c:915
> #12 0x00000000005a6acb in composition_reseat_it (cmp_it=cmp_it@entry=
>     0x7fffffffacb8, charpos=305, bytepos=322, endpos=1, w=0x1292a90, face=
>     0x3dca4e0, string=0) at composite.c:1246
> #13 0x00000000004419a8 in next_element_from_buffer (it=0x7fffffffa460)
>     at xdisp.c:8293
> #14 0x000000000043fd1a in get_next_display_element (it=it@entry=0x7fffffffa460)
>     at xdisp.c:6893
> #15 0x0000000000441be1 in move_it_in_display_line_to (it=it@entry=
>     0x7fffffffa460, to_charpos=to_charpos@entry=15588, to_x=to_x@entry=-1, 
>     op=op@entry=MOVE_TO_POS) at xdisp.c:8583
> #16 0x0000000000443f6c in move_it_to (it=it@entry=0x7fffffffa460, to_charpos=
>     15588, to_x=to_x@entry=-1, to_y=1040, to_vpos=to_vpos@entry=-1, 
>     op=op@entry=10) at xdisp.c:9174
> #17 0x00000000004446b6 in move_it_vertically (it=0x7fffffffa460, 
>     dy=<optimized out>) at xdisp.c:9555
> #18 0x000000000046fcb9 in Fwindow_end (window=<optimized out>, 

Thanks.  Does this happen right when you start Emacs?  Does this
happen if you start it as "emacs -Q"?

This looks like a crash in displaying some complex script, so can you
tell what text was supposed to be displayed?

Also, if this started happening lately, can you bisect to find the
commit which exposed this problem?

Thanks.